EATING THE HUMBLE PIE a NON DEFENSIVE RESPONSE TO FAILURES

EATING THE HUMBLE PIE: A NON-DEFENSIVE RESPONSE TO FAILURES AUDREY NG NATIONAL UNIVERSITY OF SINGAPORE 2012 EATING THE HUMBLE PIE: A NON-DEFENSIVE RESPONSE TO FAILURES AUDREY NG (B.S.Sc (Hons), NUS) A THESIS SUBMITTED FOR THE DEGREE OF MASTER OF SOCIAL SCIENCES DEPARTMENT OF PSYCHOLOGY NATIONAL UNIVERSITY OF SINGAPORE 2012 i DECLARATION I hereby declare that the thesis is my original work and it has been written by me in its entirety. I have duly acknowledged all the sources of the information which have been sued in the thesis This thesis has also not been submitted for any degree in any university previously. __________________________ Audrey Ng Shuhui 27 February 2013 ii ACKNOWLEDGMENTS This thesis had been a major undertaking in my graduate days and I have the following to thank for the successful completion of it: Dr. Eddie Tong, for his guidance; The ISM students, for their data collection and data entry And, all my friends. iii TABLE OF CONTENT PAGE DECLARATION ACKNOWLEDGMENTS TABLE OF CONTENT SUMMARY LIST OF TABLES i ii iii iv v CHAPTER ONE INTRODUCTION a. Nature of Humility b. Humility: A Character Strength c. Humility: Non-Defensive Response to Failure d. Current Research and Methodology Considerations 1 1 4 6 9 CHAPTER TWO EXPERIMENT ONE a. Overview b. Method c. Results d. Discussion 12 12 15 19 CHAPTER THREE EXPERIMENT TWO a. Overview b. Method c. Results d. Discussion 21 23 26 31 CHAPTER FOUR EXPERIMENT THREE a. Overview b. Method c. Results d. Discussion 34 35 36 40 CHAPTER FIVE GENERAL DISCUSSION a. Reconciling the findings with Literature b. Further Implications and Conclusions 41 43 45 REFERENCES 52 iv SUMMARY Humility has been classified as a character strength and has been thought to bestow positive life outcomes. This research aimed to examine the effects of humility in response to failures. I hypothesized that humble individuals cope with failures in a non-defensive manner. Across three experiments, it has been demonstrated that after receiving negative feedback individuals induced to feel humble do not show over-activation of positive self-concepts neither do they displayed a diminished activation of negative self-concepts compared to the non-humbled individuals. That is, they would not deliberately magnify their merits nor diminish their weaknesses in order to cope with an ego-threat. How these findings reconcile with the nature of humility is discussed. Comparisons between humility and low and high self-esteem were also reviewed. v LIST OF TABLES PAGE TABLE 1 Percentage of Humility Traits in Participants Recalled Descriptions (Experiments 1 to 3) 47 TABLE 2 Proportion of Positive Traits listed across conditions (Experiment 1) 48 TABLE 3 Response Tendencies: Mean Number of self versus nonself selections across conditions (Experiment 2 & 3) 49 TABLE 4 Means of Reaction Time to Positive and Negative Self-concepts across conditions (Experiment 2) 50 TABLE 5 Means of Reaction Time to Positive and Negative Self-concepts across conditions (Experiment 3) 51 1 CHAPTER ONE INTRODUCTION Research on human character strengths and virtues include the study of humility. Humility has been classified as a character strength (Peterson & Seligman, 2004) and has been thought to bestow positive life outcomes (e.g., Emmons, 1999; Sandage, 1999; Sandage & Wiens, 2001; Worthington, 1998). Although the benefits of humility have been proposed by both psychologists and religious leaders alike, the study of humility and its effects on basic psychological effects still remain limited. In particular, questions remains regarding how humble people respond to failure that threatens the self (i.e., egothreat) because of the personal shortcoming it communicates. Research has found that the manner in which a person responds to failure has implications for subsequent adjustment and well-being. If humility is a character strengths that offers resources needed for effective coping, does it also enable a constructive response to ego-threats? In this research, I aim to fill a gap in humility research by examining the effects of humility in face of failures. Nature of Humility The study of humility can be traced back to the teachings of theology and philosophy, many of which expound the merits of being humble (e.g., Richards, 1992). Dictionary definitions tend to portray humility negatively and associate it with low selfregard and unworthiness. For example, the Longman Dictionary defines humility as considering the self as less important than others whereas the Oxford English Dictionary defines humility as having a low view of one’s importance. Note that although these definitions are confined to the layperson’s perception, studies have found exceptions. For 2 example, in a study by Exline and Greyer (2004), the researchers found that American college students perceive humility as a psychological strength and do not equate humility with low self-esteem. Still the academias painted a more favorable picture and conceptualized humility much more accurately. In particular, humble individuals are said to have an assured sense of self-worth (Tangney, 2002; Exline & Greyer, 2004). Psychologists have posited that this assured sense of self-worth entails humble individuals knowing their strengths and limitations. This does not mean that humble people know themselves perfectly well. Rather, it implies that humble individuals are willing to see themselves accurately, to know both their strengths and weaknesses, and they would not deliberately magnify their merits nor diminish their weaknesses (Means, Wilson, Sturm, & Biron, 1990; Peterson & Seligman, 2004; Exline & Greyer, 2004). Consistent with their strong sense of self-worth, experts have proposed that humble individuals are able to acknowledge and accept their shortcomings (Sandage, 1999; Emmons, 1999; Jennings et al., 2005). They are thought to be willing to acknowledge and accept information of the self even if they are negative, and would not reject or distort this negative information so as to protect their self-image (Tangney, 2000). As noted by Yancey (2000), one reason why early Christian writers are often perceived as uniquely humble is their ability to acknowledge their mistakes and shortcomings. Consistently, Myers (1995, 2000) observed that humble people exhibit very little self-serving bias and tend not to take extra credit for their accomplishments. These observations reiterate that humble people are assured individuals who tend not to 3 magnify their strengths or ignore their weaknesses to feel better about themselves (Means et al., 1990; Peterson & Seligman, 2004; Exline & Greyer, 2004). One need to note that humility is not the same as self-deprecation in which the self is disparaged, often for self-serving reasons (e.g., to lower expectations, to seek social approval). A study by Gibson (2007) demonstrated that individuals who selfdeprecate are not interested in having an accurate sense of self. Humility is also not modesty. A modest person behaves in a socially approved manner, such as not dressing in a flashy manner or not taking credit for a success, but may privately have a sense of superiority (Hineline, 1991; Peterson & Seligman, 2004). A humble person, on the other hand, behaves in accordance to his/her measured self-views and resists seeing himself/herself as more important than others. Humble individuals are also known to be not arrogant for these reasons. They are thought to be sensitive and responsive to the feelings of other people (Exline, Baumeister, Bushman, Campbell, & Finkel, 2004; Means et al., 1990; Rowatt et al., 2006; Tangney, 2000). They respect the worth of each individual, and do not put people down (Means et al., 1990; Roberts, 1983; Worthington, 1998). In fact, studies have found that college students who are humble in their achievements are preferred by their fellow schoolmates (Hareli & Weiner, 2000; Sedikides, Gregg, & Hart, 2007). The fact that humble people do not self-enhance or self-deprecate could also make them more likeable and respected in the eyes of others. Further, because humble people are not defensive about their weaknesses and have the desire to improve themselves, they are often thought to be open to correction (Tangney, 2002; Templeton, 1998). They are ready to change themselves or make 4 amendments for their mistakes (Davis, Worthington, & Hook, 2010; Exline et al., 2004; Means et al., 1990; Rowatt et al., 2006). Theorists have also noted a strong sense of spirituality/religiosity in humble people (Powers et al., 2007; Rowatt et al., 2006; Tangney, 2002). The strong religious beliefs that humble people are thought to hold might be a channel by which they learn about themselves, keep their achievements in perspective, acknowledge and improve upon on their weaknesses, and maintain a respectful and non-boastful approach in their relationships with others. All in all, humility can be thought as a dispositional character strength that brings about positive life outcome, whether directly or indirectly. To be accurate, research has yet to thoroughly validate which of the mentioned qualities of humility indeed constitute the nature of humility and which qualities are simply related to the disposition. This ambiguity is in part due to the lack of validated humility measures (which I will explain shortly). However, drawing from these scholastic conceptualizations as a starting point, I propose that a working definition of humility could involve the following attributes: (1) an assured self-worth; (2) tendency to keep one’s abilities and accomplishments in perspective; (3) willingness to acknowledge and accept one’s mistakes and imperfections; (4) lack of self-deprecating tendencies; (5) lack of self-enhancing tendencies; (6) willingness to improve on one’s weaknesses. One need to note that humility should be considered as a dynamic construct comprising these attributes, hence they may not be overlapping perfectly. Humility: A Dispositional Character Strength Many researchers have argued that one reason that humility is important is because it predicts positive life outcomes (e.g., Emmons, 1999; Peterson & Seligman, 5 2004; Sandage, 1999; Sandage & Wiens, 2001; Worthington, 1998). There is evidence to suggest that humble individuals are forgiving, compassionate, patient, and kind (Davis et al., 2010; Exline & Geyer, 2004; Means et al., 1990; Powers et al., 2007; Peters et al., 2011; Rowatt et al., 2006; Sandage, 1999). For instance, in a study by Exline and colleagues (2000) it was found that people who were in a state of humility were slower to retaliate in response to provocation on a laboratory task. This implies that humility could positively impact interpersonal relationships. Further, the willingness to acknowledge and accept one’s mistakes and imperfections (considered to be one of the attributes of humility) may result in decreased justification of their actions. That is, they will not blindly defend themselves and/or their actions. While defense and justification of one’s actions will result in a perception of injustice, decreased justification will instead promotes the seeking and giving of forgiveness (Exline et al., 2004). People who forgive have even been shown to reap physiological benefits (Lawler et al., 2003; Lawler et al., 2005). There are also studies which found a link between humility and good academic performance (Johnson, Rowatt, & Petrini, 2011; Rowatt et al., 2006). Though the causality of this correlation is unclear, it is consistent with proposals that humble individuals are open to novel ideas and have a strong desire to learn (Tangney, 2002; Templeton, 1998). Being open and having a strong desire to learn is also a desirable trait in the work setting (Reave, 2005). It is thus possible that humble people are well-liked and respected whether in school or work setting, and that people are therefore willing to assist them in times of need (Peters, Rowatt, & Johnson, 2011). Also, humble corporate leaders who are resilient were also found to generate more profits for the organization 6 (Collins, 2001), while humility in employees was found to predict high job performance (Johnson et al., 2011). Humility had also been shown to be beneficial in health and therapy settings (Kurtz & Ketcham, 1992; Fontana, Rosenberg, Burg, Kerns, & Colonese, 1990). Though the causality of these findings is unclear, they could be tentative support for the idea that humility is a resource that enables effective adaptation to life’s challenges (Emmons, 1999; Exline & Geyer, 2004; Petersen & Seligman, 2004; Tangney, 2002). Humility: Non-Defensive Response to Failure The literature suggests that humble individuals adapt well to life events, whether they involve mastering academic concepts, managing relationships, or handling work challenges. However, navigating such life events is not easy because of the intermittent failures one could encounter, and effective adaptation to challenges depends in part on how the person responds to failures. Of interest in the current research is how humility will influence the effect that failure could have on self-concepts. A body of research has shown that failure can be damaging to a person’s self-views and also their subsequent performance (Bronckner, 1979; Morrison, 1979; Nurius & Markus, 1990). Failure can signal inadequacies in the self and some people react by affirming their positive selfviews whereas other might respond by wallowing in self-pity (Neff, Hsieh, & Dejitterat, 2005). If humility is associated with an assured sense of self-worth, what effect would failure have on the self-concepts of the humble? This research examined how selfconcepts are activated during a failure encounter as a function of humility. Failures, such as doing badly at a term paper or being rejected at an interview, are inevitable in life, but people react to them differently. To cope with the ego threat that 7 follows failures, one can self-enhance by inflating positive self-views, deflating negative self-view or both (Feick & Rhodewalt, 1997; Sedikides & Strube, 1997; Arkin, 1981). For example, a student who has obtained an undesired grade may try to discredit the negative feedback by deliberately attributing his or her failure to external factors (e.g., the room is too noisy for me to concentrate) rather accepting his or her weaknesses (Blaine & Crocker, 1993). The student might also try to undermine the failure by bringing to mind positive self-concepts (e.g., remembering that he is good in sport) and rejecting negative self-concepts (e.g., discounting the fact that he has been lazy; Baumeister & Jones, 1978; Steele, 1988). Using these self-enhancing coping strategies, the student could disconfirm the unfavorable feedback received in the failure, maintain or even inflate her/her sense of self-worth, and deny personal weaknesses. Self-enhancing is a common strategy to cope with failures, for instance, Steel (1988) suggested that people may compensate for a failure experience by focusing on their strengths. In a study by Baumeister and Jones (1978), it was found that after receiving a negative feedback, participants gave themselves higher rating for selfattributes that were unrelated to the failure. It was also found that favorability of selfimage increased when participants encountered a failure in public (Greenberg & Pyszczynski, 1986). This suggests that an overvaluation of self-image is a way to compensate for failures. Besides these empirical support, self-enhancing defenses have long been suggested by Adler (1956) who theorized that the strive for superiority is an attempt to compensate for any inferiority. However, there are grounds to propose that humble individuals are less likely to employ such self-enhancing strategies to cope with ego threats. As just reviewed, there 8 appears to be consensus among theorists that humility is associated with an assured sense of self-worth. Compared to the average person, humble people are more willing to accept their shortcomings and to keep their accomplishments in check. This could suggest that in the event of a failure, they have less compulsion to inflate their self-worth by magnifying positive self-information and rejecting negative self-information to disconfirm the undesirable failure feedback. This could stem not only from their secured sense of selfworth but also from their willingness to accept their weaknesses and having an openness to correct themselves (Sandage, 1999; Tangney, 2002; Exline et al., 2004). Note also that humble individuals are aware of their strengths. These strengths can help them to buffer against failures in life. It is thus possible that during failures, humble individuals would draw on their strengths and would be less averse to the negative feedback about themselves, less likely to reject the negative self-information, and more likely to learn from the experience (Tangney, 2002; Means et al., 1990). Therefore, I predict that upon receiving negative feedback signifying failure, humble individuals are less likely than non-humble individuals to over-activate positive self-concepts and also less likely to de-activate negative self-concepts. This nondefensive strategy, if true, could yield psychological dividends for humble people. Studies have found that people with inflated self-views are more socially maladaptive and less liked by their peers (Colvin, Black, & Funder, 1995; Robinson, Johnson, & Shields, 1995). Further, research by Moore (2007) found that positive illusions correlated positively with depressive symptoms. Extremely positive self-views have also been linked to poorer self-esteem, well-being, and even academic performance (Compton, 2001; Robins & Beer, 2001). Hence, an contribution that this research could make for 9 future research on humility is in showing how the non-defensive response to failures exhibited by humble individuals might facilitate coping with difficulties and excelling in academic, relationship, and work challenges (Johnson et al., 2011; Rowatt et al. 2006). Current Research and Methodological Considerations The current research examined differences in the type of the self-concept activated in response to a failure situation between humble and neutral (non-humble) people. To simulate a failure situation, some participants were asked to do an arithmetic task and were given feedback that they had done poorly. Other participants completed the same task but were not given any feedback. Self-concepts were then measured. Note that this research examined only global positive and negative self-concepts (good versus bad self-related information) and not domain-specific self-concepts (e.g., intelligence). Activation of global self-concepts was operationalized and measured differently in different studies. In Experiment 1, they were represented by the number of positive and negative traits of themselves participants can bring to mind. In Experiments 2 and 3, we employed a latency response task to measure accessibility of global positive and negative self-concepts. My over-arching prediction is that in the negative feedback condition, humble participants would exhibit weaker activation of negative self-concepts (Experiment 1: fewer negative traits listed; Experiments 2 and 3: weaker accessibility of negative self-concepts) as well as weaker activation of positive self-concepts (Experiment 1: fewer positive traits listed; Experiments 2 and 3: weaker accessibility of positive selfconcepts) compared to neutral participants. This prediction could be tested by measuring humility and examining how selfconcepts varied with humility. However, there are formidable validity issues with 10 available measures of humility. Self-report humility scales are not suitable for several reasons (e.g., Ashton & Lee, 2005; 2007; 2008). Genuinely humble people would not report their humble qualities as they are less attentive to themselves (Tangney, 2000; Myers, 1995). Further, people could report themselves as highly humble, a desirable quality, due to self-serving motivations (Johnson & Robins, 1993; Asendorpf & Ostendorf, 1998). This critique should not be taken to imply that humble and non-humble individuals never provide accurate scores of their humility. Rather it is the acknowledgment that the current self-report measures are not able to provide an accurate, holistic and reliable measure of humility and that users of the self-report humility measures should be aware that such limitations. Further, an implicit association test (IAT) of humility has also been developed to circumvent concerns associated with self-report (Powers et al., 2007; Rowatt et al., 2006). 1995). However, this measure failed to correlate with humility peer ratings and other humility-related outcomes, raising concerns over its validity (Powers et al., 2007; Rowatt et al., 2006). Therefore, I manipulated humility and tested differences in activated self-concepts between humility and neutral conditions. However, past studies have not attempted to manipulate humility. Humility could be manipulated by staging realistic situations or having participants imagine themselves in suitable vignettes or videos. However, the same situation or imagery material might not generate the intended response in all participants; e.g., participants may feel low self-esteem instead of humility (HarmonJones, Peterson, & Vaughn, 2003). Implicit methods (e.g., subliminal priming) could be used, but they tend to activate implicit representations, not necessarily conscious 11 experiences (Winkielman, Berridge, & Wilbarger, 2005; Yang & Tong, 2010), and my aim was to examine conscious humility experiences. The recall method was used, in which, before the failure simulation, participants (in all three studies) were asked to recall an incident where they felt humble. Specific definitions of humility were given to the participants such that they would not confuse humility with low self-regard and humiliation. One other humility study by Exline and Geyer (2004) also had participants recall humble experiences, although not for the purpose of inducing humility. Recall is an effective method for inducing psychological states (Gerrards-Hesse, Spies, & Hesse, 1994). There are concerns of memory degradation with the recall method, but it avoids or minimizes the problems of other induction techniques. Participants in the neutral condition were asked to indicate the landmarks they encountered while commuting to school. They were not asked to recall ‘neutral events’ so as to avoid participants self-selecting the events and consequently recalling events with varying emotional tone. I also included a low self-esteem and high self-esteem condition in Experiments 2 and 3 respectively. Laypersons might equate humility to low self-worth, a trait synonymous with low self-esteem. The notion that humility involves an assured sense of self might make it seems too similar to high self-esteem. As such, humility might be confused with both low self-esteem and high self-esteem. However, these constructs are different and it is necessary to conceptually and empirically differentiate humility from low and high self-esteem by showing that they produce different effects. Evidence in this regard would assist in establishing humility as a distinctive virtue (Peterson & Seligman, 2004). 12 13 CHAPTER TWO EXPERIMENT 1 Overview In Experiment 1, participants were asked to list down traits about themselves after the humility and feedback manipulation. Previous research found that in general, positive attributes of the self becomes more accessible than negative attributes during failures as part of the compensatory strategies to eliminate ego-threat (Wood, Giordano-Beech, & Ducharme, 1999; Steele, 1988). Hence, it is possible that participants in both neutral and humility conditions, not just the neutral condition, would bring to mind more positive traits, and also fewer negative traits, when given negative feedback compared to when they were not given any feedback. In addition, it is proposed here that humility weakens the need to engage self-enhancing strategies to cope with ego threats. Hence, I predicted that when given negative feedback, participants in the humility condition should list down a lower proportion of positive traits (i.e., higher proportion of negative traits) than participants in the neutral condition. Method Participants Sixty-three undergraduates (52 females; M = 20.36, SD = 1.47) from the National University of Singapore (NUS) participated for course credits. Participants were randomly assigned to one of the four conditions in a 2 (condition: humility versus neutral) × 2 (feedback: negative feedback versus no feedback) factorial design. See Table 2 for cell sizes. Gender was not analyzed as a variable across all studies because of the disproportionately lower number of males. 14 Procedure The experiment was carried out in groups of 10 in partitioned cubicles to ensure anonymity and privacy. Upon arrival, participants were informed that they would be performing several unrelated experiments. The first experiment was presented as a “Scale Validation Study” which was a cover for the humility manipulation. Participants in the humility condition were asked to recall a humble experience. They were encouraged to describe as much details as possible so that anyone reading their description could understand their experience. Question prompts (“please indicate how you felt”, “how it feels like to be feeling humility”) were given to facilitate recall. Participants were instructed not to confuse humility with humiliation, feelings of shame, and low self-worth. Definition of humility was also provided, detailing humility as being aware of one’s weaknesses, modest of achievements, and acceptance of the fact that there are others who are better. In contrast, participants in the neutral condition were asked to describe the landmarks (e.g., stations, schools) they encounter while commuting to the school and their daily routine on the evening before they go to bed. The purpose of getting the neutral participants to describe two activities was to make comparable the time both the humility and neutral participants spent on this recall task. After the recall, participants proceeded to a task that measured the current dependent variable. They were told that they would be participating in a “Conceptual Thinking Study” which would measure their ability to think conceptually. Twenty questions from the quantitative section of the Graduate Record Examination (GRE) were presented using Medialab to the participants. The questions tested logical thinking ability in a multiple-choice format and were presented one at a time. Writing materials were 15 provided for the participants to work out the answers. Participants were given up to 45 seconds to solve each question, after which the next question would appear on the computer screen regardless whether they had indicated an answer. The entire task lasted 15 minutes, after which those in the negative feedback condition were told to wait for a short while for their score. About a minute later, they were informed that they scored below the 20th percentile of their undergraduate cohort. Those in the no feedback condition were not told that their scores could be available and were not informed of their results. Next, all participants completed the trait-listing task. They were told to list down as many positive and negative traits about themselves they could bring to mind at the moment. The trait-listing task was self-paced with no time limit. Last, the participants completed manipulation checks and demographics items before they were debriefed and released. Measures Reported humility As a manipulation check for the humility manipulation, participants in all conditions rated, on a scale ranging from 1 (strong disagree) to 7 (strong agree), three items: “Do you think the situation described above reflects humility?”, “How humble do you feel right now?”, and “How meek do you feel right now?”. Scores for the items were averaged to produce reported humility (α = .90). Negative feedback As manipulation checks for the negative feedback manipulation, participants in the negative feedback condition were asked to indicate the percentile they scored. They also rated a perceived feedback accuracy item, “How 16 accurate do you perceive the score to be?” on a seven-point scale ranging from 1 (not accurate at all) to 7 (very accurate). Results Manipulation Checks Humility induction An independent samples t- test conducted on the reported humility indicated participants in the humble condition reported feeling more humble (M = 4.80, SD = .45) than those in the neutral condition (M = 2.80, SD = .93), t(61) = 10.89, p < .001, ηp2 = .68. While these findings might suggest that the manipulation was successful, they could still be of suspect considering the validity issues with self-report measures of humility. As an added form of manipulation check, I coded the humility condition participants’ recall descriptions for the extent to which they were consistent with current conceptualizations of humility. By coding whether participants’ recall descriptions coincide with theoretical conceptualizations of humility, I am assuming that the description can be an indicator, albeit an imperfect one, of how much they had been induced into a humble state. It is important to note as a caveat that not everyone who writes about humility would feel humble. However, there is considerable evidence from affect-induction research using the recall method that writing about specific affective states is generally effective at inducing the targeted states. Based on the literature review, six humility attributes were identified. They appear to be common across many theoretical models and empirical findings (e.g., Myers, 1979; Hwang, 1982; Means et al., 1990; Clark, 1992; Richards, 1992; Halling, Kunz, & Rowe, 1994; Templeton, 1998; Emmons, 1999; Exline & Geyer, 2004; Tangney, 2005). 17 First, humble individuals are thought to have an assured sense of self. This assured sense of self-worth would entails having an accurate sense of one’s strengths and limitations, not deliberately magnify their merits nor diminish their weaknesses (Means et al., 1990; Templeton, 1998; Tangney, 2000). Humble individuals also have the tendency to keep one’s abilities and accomplishments in perspective. This would mean not taking take extra credit for accomplishments and not magnifying their strengths to inflate their selfworth (Means 1995, 2000; Peterson & Seligman, 2004; Exline & Greyer, 2004). Further, they are willing to acknowledge and accept their mistakes and imperfections (Clark, 1992; Emmons, 1999; Jennings et al., 2005). The lack of self-deprecating tendencies is also a prominent attribute of the humble. For instance, Gibson (2007) found that individuals who self-deprecate are not interested in having an accurate sense of self. This runs contrary to what is known about humble individuals; their willingness to see themselves accurately. In addition, humble individuals also lack of self-enhancing tendencies (Myers, 1995; 2000; Means et al., 1990; Richards, 1992). Last, the humble are willing to improve on one’s weaknesses (Sandage, 1999; Jennings et al., 2005). Every humility recall description was independently coded by two coders for the presence (coded as 1) or absence (coded as 0) of each attribute. That is, each humility description could obtain a minimum total score of 0 (i.e., it contained none of the six attributes) to the maximum total score of 6 (it contained all attributes). Statements of the six attributes as expressed in Table 1 were given to the two coders. One of the coders is blind to the purpose of the experiment. Any misconception and disagreement about the meaning of each attribute was first discussed and resolved. The coders then proceeded with a few descriptions as practice. It was highly unlikely that any participants would so 18 clearly state that, for example, he/she had a sense of self-acceptance. The coders would need to read the entire description and make an inference of whether, in this example, the participant had exhibited a sense of self-acceptance in the situation he/she described. Such coding is unavoidably subjective and hence the need for cross-validation between two coders was all the more necessary. The coding data, in terms of the percentage of descriptions containing each attribute, are presented in Table 1. An intra-class correlation coefficient (ICC) of .84 was obtained, indicating a high level of agreement between the two coders. Any disagreements in coding were resolved and the final rating is presented in Table 1. Given that it was not always possible to code an attribute from a description, and also that not every attribute would be expressed in every instance of humility (e.g., the participant did not have to keep his/her accomplishment in perspective because there was no accomplishment in the described situation in the first place), we did not expect the proportions to be too high. However, we conducted a one-sample t-test on the proportion of each of the attribute against 50% and found that all but “Lack of self-deprecating tendencies” is significantly above 50%. This implies that at least 50% of the participants’ recall descriptions contained these attributes. Further, assuming that a recall describing humility should contained at least 50% of the humble attributes, I conducted another onesample t-test comparing the agreed rating against 50% (i.e., at least 3 out of 6 of the humble attributes) and found that the recall descriptions contained a significantly higher number of humble attributes, t(31) = 2.88, p = .007, ηp2 = .21. Note that there was no data to compare these proportions with because the neutral condition was asked about landmarks on the way to school which could not be coded. It 19 was therefore not clear whether non-humility descriptions would also contain these attributes. This issue would be pursued further in Studies 2 and 3 where the proportion of humility attributes were compared with that in two non-humility constructs, low selfesteem and high self-esteem, respectively. Negative feedback induction. Combining the humility and neutral conditions, participants in the negative feedback condition on average indicated that they scored on the 17.52th (SD = 2.94) percentile of the cohort. Hence, the negative feedback participants in general were accurate in recalling their feedback. Also, there was no significant difference between the humility participants (M = 17.25, SD = 2.69) and the neutral participants (M = 18.0, SD = 3.46) in the negative feedback condition in terms of the percentile feedback they indicated they received, t(24) = .65, p = .53, ηp2 = .57 . Further, across the humility and neutral conditions, the average perceived feedback accuracy score in the negative feedback condition was 4.88 (SD = .82). Noting that perceived feedback accuracy ranged from 1 (not accurate at all) to 7 (very accurate), this suggests that the negative feedback participants in general found the feedback to be a moderately accurate. A t-test revealed no effect of condition on perceived feedback accuracy, t(24) = .99, p = .33, ηp2 = .04 . Main Analysis The numbers of positive and negative traits listed in each condition are presented in Table 2. In order to control for the differences in the total number of positive and negative traits listed, the proportions of positive and negative traits out of the total number of traits were computed and analyzed as dependent variables. As the proportions 20 for the positive and negative traits were opposite reflections of each other, only the proportion of positive traits was analyzed and discussed. A 2 (condition) × 2 (feedback) ANOVA conducted on the proportion of positive traits (see Table 2 for the averaged proportions). The analysis revealed a significant main effect of condition such that participants in the neutral condition listed more positive traits than those in the humility condition, F(1,59) = 4.11, p = .047, ηp2 = .07. There was a significant main effect of feedback such that participants in the negative feedback condition listed a higher proportion of positive traits than participants in the no feedback condition, F(1,59) = 5.53, p = .022, ηp2 = .09. However, the expected interaction effect was not significant, F(1,59) = .003, p = .95, ηp2 < .001. Despite the non-significant interaction, I proceeded with simple effect analyses. Post-hoc analyses using pair-wise comparisons revealed that when given negative feedback, participants in the neutral condition listed more positive traits than participants in the humility condition, t(31) = 3.69, p = .001, ηp2= .31 (see Table 2). This trend was not observed in the no feedback condition, t(26) = 1.42 p = 0.68, ηp2 = .07 . Also, negative feedback led to a significantly higher proportion of positive traits than no feedback, in both the humility condition, t(30) = 2.20, p = .04, ηp2 = .14, and the neutral condition, t(29) = 3.29, p = .003, ηp2 = .27. Discussion Experiment 1 provided the preliminary support for the current hypotheses. After receiving negative feedback, participants in both neutral and humility conditions listed more positive traits and less negative traits, than without feedback. Hence, there was evidence to suggest that both neutral and humility participants engaged in compensatory 21 strategies during a failure situation by bringing to mind more positive thoughts of themselves. There was also some evidence to suggest that this compensatory effect was weaker among the humility participants. When given negative feedback, participants in the humility condition listed fewer positive traits than those in the neutral condition. This pattern was not found when participants were not given any feedback. However, the interaction effect was not significant, and the evidence presented here could only be considered tentative. Despite the promising results from Experiment 1 there was a need to replicate and substantiate these preliminary findings. By asking participants to self-report the positive and negative self-concepts they can bring to mind, the responses may be susceptible to self-presentation motivations. Participants may not respond truthfully due to social desirability or impression management. In Experiment 2, a response latency paradigm is used to circumvent the limitation of using self-report. Also, by analyzing the response latency of both the positive and negative traits separately I will able to examine the differential accessibility of positive and negative traits. An additional low self-esteem condition was added for comparison. 22 CHAPTER THREE EXPERIMENT 2 Overview Experiment 2 was conducted for three reasons. First, it aimed at obtaining statistically stronger results. The second objective is to substantiate earlier findings by measuring accessibility of positive and negative self-concepts using response latency (i.e., reaction time, RT). The accessibility of self-concepts in associative networks can be measured by the response speed towards corresponding stimuli (Higgins, 1996). The presence of a stimulus representative of a self-concept would be responded to more quickly if the self-concept is more accessible. Hence, response latency provides an indication of the accessibility of the associated concepts and affords an investigation of the differential in accessibility of positive and negative self-concepts as function of humility in a failure context. To this end, participants completed a task designed by Markus and Kunda (1986) to measure accessibility of implicit self-concepts in which they responded “me” or “not me” to a series of positive and negative trait words. Differences in reaction time (RT) to the positive and negative trait words as a consequence of the same humility and feedback manipulation would be examined. The third objective was to differentiate humility from low self-esteem (LSE) by examining whether they engender different accessibility of positive and negative selfconcepts in a failure situation. As noted, some dictionaries portray humility negatively by equating it low self-esteem (e.g., Longman Dictionary). However, there are marked differences between humility and LSE. LSE individuals have a tendency to compound the effects of their failure by increasing negative thoughts regarding themselves and 23 overgeneralizing their weaknesses to unrelated domains (Kernis, Brockner, & Frankel, 1989; Brown & Dutton, 1995). They would not only accept negative feedback regarding themselves in the event of a failure, but would also evaluate themselves very negatively thereafter (Shrauger, 1975; Shrauger & Rosenberg, 1970; McFarlin & Blascovich, 1981). In contrast, as argued, humble individuals might accept the negative feedback but they should be less critical of themselves. Also, it is posited here that they are less likely to trigger a high amount of negative self-concepts following a failure. To examine whether humility and LSE produce different effects, LSE was also manipulated, using the same recall technique. Taking into consideration the findings from Experiment 1 and research on LSE, I hypothesized that with negative feedback, the neutral condition should display the shortest RT to positive trait words followed by the humility condition, and LSE condition should exhibit the longest RT. A reverse trend was hypothesized for the negative trait words; i.e., when given negative feedback, the neutral participants should show the longest RT followed by humility participants, with LSE participants displaying the shortest RT. In addition, it was hypothesized that participants in both neutral and humility conditions would respond faster to the positive trait words, and slower to the negative trait words, when given negative feedback compared to when they were not given any feedback, whereas participants in the LSE condition would respond slower to the positive trait words, and faster to the negative trait words, in the negative feedback condition as compared to the no feedback condition. 24 Method Participants 185 NUS undergraduates (135 females; mean age = 20.0; SD – 1.56) participated for partial course credits. The participants were randomly assigned to one of the six conditions in a 3 (condition: humility versus neutral versus low self-esteem [LSE]) by 2 (feedback: negative feedback versus no feedback) factorial design. Cell sizes are indicated in Table 4. 9 participants failed the manipulation checks and were removed (to be described further in Results). Procedure As in Experiment 1, participants completed all tasks in isolated cubicles. The first task was the same recall task used in Experiment 1 but with an additional LSE condition. Participants in this condition were asked to recall an incident which affected their selfesteem in a negative way. This was defined to them as an incident which made them had negative evaluations of themselves and low self-esteem was further defined as a state of worthlessness. As in the humility and neutral conditions, they were encouraged to describe as much details as possible, and were furnished with similar question prompts. They then completed the “Conceptual Thinking Study” as in Experiment 1. Next, participants completed the self-concept accessibility task. Accessibility of positive and negative self-concepts was measured using response latency. Participants were told that their self-perception will be measured. The whole self-perception task including the instructions was administered using the E-prime program. The instructions explain that on each trial, they would be presented with a trait and they had to indicate whether the trait was representative of them by pressing specific keys (one marked as self and the 25 other marked as “non-self”) on the button box. Detailed instructions were given to emphasize the importance of responding as quickly as possible while being accurate with their response. Participants were asked to place their left and right index finger on the specific buttons before the task commenced. In this task, participants went through a block consisting of 60 main trials for 2 times. The block consists of 20 positive traits, 20 negative traits and 20 neutral traits. The presentation of the trials within the block was randomized across participants. For each trial, a fixation cross (+) was presented at the centre of the screen for 1000ms, followed by either the positive trait, negative trait or neutral word. These traits/words were presented for 3000ms. Response to the trait/word was self-paced but if a response is not made after 3000ms, it would skip to the next trait/word. A 1000ms break was given between each block. This task was adapted from Markus and Kunda (1986) where participants need to respond “me” or “not me” to stimuli relating to specific traits. Finally, the participants completed manipulation checks and demographics items before they were debriefed and released. Measures Reported humility Participants in all conditions rated the same reported humility as in Experiment 1 (α = .83). Negative feedback Participants in the negative feedback condition were also asked the same percentile question and rated the same perceived feedback accuracy item. State self-esteem Participants in all conditions completed a modified version of state self-esteem scale (Heatherton & Polivy, 1991) consisting of performance and social factors that are thought to be sensitive to changes in different aspects of selfconcepts. The scale using a five-point Likert scale that ranged from 1 (not at all) to 5 26 (extremely). The appearance component was removed because it was found to be relatively stable in face of laboratory or academic failures (Heatherton & Polivy, 1991). Also, this factor is irrelevant to the current failure design of the experiment. Self-concept Accessibility Task Using frequency of appearance, the traits given by participants from Experiment 1 were ranked. Based on the ranking, I generated a list of 40 traits (20 positive and 20 negative). The positive traits used were “easy-going”, “thoughtful”, “knowledgeable”, “selfless”, “determined”, “hardworking”, “friendly”, “patient”, “trustworthy”, “respectful”, “loyal”, “sociable”, “organized”, “helpful”, “responsible”, “compassionate”, “strong-willed”, “forgiving”, “conscientious”, and “confident”. The negative traits used were “tactless”, “disorganized”, “procrastinator”, “immature”, “ lack of confidence”, “insensitive”, “cynical”, “un-forgiving”, “stubborn”, “hot-tempered”, ”careless”, “selfish”, “messy”, “harsh”, “lazy”, “non-resilient”, “unmotivated”, “impatient”, “un-expressive”, and “lack of sympathy”. In addition, I also generated a list of neutral words consisting of “structured”, “vertical”, “advanced”, “horizontal”, “basic”, “circle”, “pointed”, “floating”, “curved”, “transparent”, “gravity”, “translucent”, “frontal”, “concave”, “linear”, “fundamental”, “angled”, “cloudy”, “dormant”, and “opaque”. This list of traits/words were then checked using the English Lexicon Project (ELP; see http://elexicon.wustl.edu/) database to examine the lexical characteristics. Only words that were fairly equivalent in terms of usage frequency and word length were used.1 1 The three word types did not differ in usage frequency, F(2,57) = 2.58, p = .08, ηp2 = .08, and word length, F(2,57) = 1.79, p = .18, ηp2 = .06. 27 Results Manipulation Checks Humility induction An one-way ANOVA revealed a significant effect of condition on reported humility, F(2,173) = 17.50, p < .001, ηp2 = .17. Independent samples t- tests indicated that participants in the humility condition reported feeling more humble (M = 4.59, SD = .66) than those in the neutral condition (M = 3.16, SD = 1.27), t(111) = 12.37, p < .001, ηp2 = .58, and those in the LSE condition (M = 3.21, SD = 1.12), t(116) = 9.48, p < .001, ηp2 = .44, and no difference between the neutral and LSE conditions, t(119) = .21, p = .836, ηp2 < .001. As in Experiment 1, the presence/absence of the same six humility attributes in the descriptions was independently coded by two coders. Recalled descriptions from both humility and LSE conditions were coded, so that the extent to which the attributes were present in both types of descriptions could be compared. A high degree of inter-rater reliability was obtained, intra-class correlation coefficient (ICC) = .93. As shown in Table 1, the attribute proportions in the humility condition were comparable to those obtained in Experiment 1. All of the attributes were significantly above 50%, implying that at least half of the participants’ recall descriptions contained the attributes. In contrast, none of the recall descriptions in the LSE condition contained any of the humble attributes hence comparisons between both conditions were not possible. I then conducted another one-sample t-test comparing the agreed rating against 50% and it was found that the recalled humility descriptions contain at least 50% of the humility traits, t(54) = 3.25, p = .002, ηp2 = .16. 28 Induction of Low self-esteem A one-way ANOVA revealed a significant effect of condition on state self-esteem, F(2,173) = 7.41, p = .001, ηp2 = .08 . Further independent samples t- tests showed that participants in LSE condition (M = 2.69, SD = .48) were lower in state self-esteem compared to those in the humility condition (M = 3.04, SD = .56), t(116) = 3.63, p < .001, ηp2 = .10, and those in the neutral condition (M = 3.05, SD = .69), t(119) = 3.30, p = .001, ηp2 = .08. No difference in state self-esteem was found between the humility and neutral conditions, t(111) = .07, p = .95, ηp2 < .001. Negative feedback induction 9 negative feedback participants (3 from the humility condition, 5 from the neutral condition and 1 from the LSE condition) incorrectly recalled their scores to be more than the 20th percentile and were removed from the analysis. All other negative feedback participants correctly recalled their feedback score. Average perceived feedback accuracy score in the negative feedback condition (across the humility, neutral, and LSE conditions) was 5.28 (SD = 1.10). Hence, the negative feedback participants in general found the feedback to be moderately accurate. A one-way ANOVA revealed no effect of condition on perceived feedback accuracy, F(2.92) = .16, p = .85, ηp2 < .001 . Main Analysis Before analyzing the latency response, the concern of whether the number of self and non-self responses differed across the six conditions and between the positive and negative traits should first be addressed. Such differential responding could in turn account for differences in latency responses. For example, slower responses to the positive trait words by humility participants as compared to the neutral participants in the feedback condition might not imply weaker accessibility of positive self-concepts in the 29 humility participants but could be due to the lower relevance of these positive traits to the humility participants (put differently, the humility participants indicated less self responses and more nonself responses to the positive trait words than the neutral participants). To address this concern, I subjected the mean number of self responses (i.e., positive-self and negative-self) in a 3 (condition: humble versus neutral versus LSE) × 2 (feedback: negative feedback versus no feedback) × 2 (traits: positive versus negative) mixed ANOVA. There was no significant main effect of condition, F(2,170) = .38, p =.68, ηp2 = .004, and feedback, F(1,170) = 1.00, p =.32, ηp2 = .004. The interactions were also not significant: condition × feedback, F(2,170) = .16, p =.85, ηp2 = .002, condition × traits, F(2,170) = .30, p =.74, ηp2 = .004, feedback × traits, F(1,170) = .007, p =.94, ηp2 < .001, and condition × feedback × traits, F(2,170) = .34, p =.71, ηp2 = .004. Note that nonself responses were not analyzed as they are reverse reflections of the analyzed indexes. This indicates that participants across the conditions have similar patterns of response tendencies. That is, the number of “positive-self” and “negative-self” responses do not differ significantly between the conditions (refer to Table 3). I also examined the response latency of self and nonself responses. Following standard procedure in analyzing reaction time (RT) data, the raw RT data (in milliseconds) was subjected to a logarithm transformation because of the positive skew in reaction time data (Tabachnick & Fidell, 1989). All subsequent analyses were conducted on these logged scores but the raw RT will be presented for ease of understanding. It was found that self responses (across both positive and negative traits) were faster (M = 938.66) than nonself responses (M = 1038.39), t(175) = 12.60, p = .007, ηp2 = .47. This finding is 30 consistent with past literature that negative or nonself responses will take longer (Fazio, 1990; Dodgson & Wood, 1998). Though the response latency of self and nonself responses is significantly different, this finding do not pose as a problem because first, participants do not differ in their tendency to respond self and nonself across conditions and second, most participants did not select nonself to the positive traits nor self to the negative traits (an average of 90% of the responses for positive traits was self and for negative traits was nonself; refer to Table 3). Next, I examined if the response latency differs significantly across positive and negative traits. For this, I conducted a 3 (condition: humble versus neutral versus LSE) × 2 (feedback: negative feedback versus no feedback) × 2 (traits: positive versus negative) mixed ANOVA. The main effect of condition is not significant, F(2,170) = .36, p =.70, ηp2 = .004 while a significant main effect of feedback was obtained, F(1,170) = 4.82, p =.03, ηp2 = .28. The interaction of condition × feedback is significant, F(2,170) = 2.57, p =.03, ηp2 = .19. Most importantly, the three-way interaction is also significant, F(2,170) = 20.41, p < .001, ηp2 = .20. This implies that the response latency differs significantly between positive and negative traits. This supports the previous finding that nonself responses are significantly slower than self responses. As an average of 90% of the responses for the negative traits were nonself and for the positive traits were self, the response latency between positive and negative traits will clearly differs. As such, the accessibility of positive and negative traits were analyzed and presented separately. Accessibility of positive self-concepts The logged RT for the positive self- concepts were submitted to a 3 (condition: humble versus neutral versus LSE) × 2 (feedback: negative feedback versus no feedback) ANOVA revealed a significant main 31 effect of condition, F(2,170) = 23.54, p < .001, ηp2 = .22, a significant main effect of feedback, F(1,170) = 6.11, p = .01, ηp2 = .04, and a significant interaction, F(2,170) = 13.01, p < .001, ηp2 = .13. Post-hoc analyses revealed that after given negative feedback, participants in neutral condition responded faster (M = 836.1) to the positive traits then participants in humble condition (M = 931.1), t(60) = 2.15, p = .04, ηp2 = .07 while the latter responded faster to participants in LSE condition (M = 1201.8), t(61) = 5.09, p < .001, ηp2 = .30. No significant difference among the conditions was found in the no feedback condition, F(2,79) = .99, p = .37, ηp2 = .01. Further analyses revealed within neutral condition, participants who received the negative feedback responded faster to the positive traits presented than those without negative feedback, (M = 1014.4), t(56) = 4.58, p < .001, ηp2 = .27. The same trend was observed in humble condition; participants who in the negative feedback condition responded faster to positive traits presented than those in the no feedback condition, (M = 1026.2), t(53) = 2.28, p = .03, ηp2 = .09. A reverse pattern was found in the LSE condition such that participants who received the negative feedback displayed a slower response to the positive traits than participants who received no feedback, (M = 1067.6), t(61) = 2.55, p = .01, ηp2 = .10 (see means in Table 4). Accessibility of negative self-concepts The same analyses conducted on the logged RT of the negative self-concepts revealed a significant main effect of condition, F(2,170) = 27.59, p < .001, ηp2 = .25, and a significant interaction, F(2,170) = 7.40, p = .001, ηp2 = .08. It was also found in negative feedback condition, participants in humble condition (M = 1005.7) have a shorter RT to the negative traits presented than participants in the neutral condition, (M = 1210.1), t(60) = 4.90, p < .001, ηp2 = .29 32 while participants in the LSE condition displayed the shortest RT, (M = 867.4), t(61) = 3.13, p = .003, ηp2 = .14. It was also found that the conditions differed significantly even in no feedback condition, F(2,79) = 4.21, p = .02, ηp2 = .05. Pair-wise comparisons revealed that participants in LSE condition (M = 963.5) responded faster to the negative traits compared to participants in neutral (M = 1090.5), t(56) = 2.48, p = .02, ηp2 = .10 and humble condition (M = 1078.2), t(53) = 2.58, p = .01, ηp2 = .11. Further analyses revealed that within neutral condition, participants who were given negative feedback showed a longer RT than those with no feedback, (M = 1090.5), t(56) = 2.77, p = .008, ηp2 = .12. No difference was found for those in humble condition, t(53) = 1.52, p = .14, ηp2 = .04. That is, participants in negative feedback condition did not respond significantly different from those in no feedback condition. Again, a reverse trend was found within LSE condition such that participants who received negative feedback responded faster to the negative traits presented than those who did not received feedback, (M = 963.5), t(61) = 2.49, p = .02, ηp2 = .09 (see means in Table 4). Discussion Using a response latency paradigm, several results from Experiment 1 concerning the differences between humility and neutral conditions were conceptually replicated and differences between humility and LSE conditions were demonstrated. First, as predicted, when given negative feedback, participants induced to feel humility displayed lower accessibility of positive self-concepts, but higher accessibility of negative self-concepts, than that of those in the neutral condition. These differences in the accessibility of positive and negative self-concepts were consistent with the differences in the proportion of positive and negative traits listed in Experiment 1. In 33 addition, as hypothesized, negative feedback increased the accessibility of positive selfconcepts relative to no feedback among the humility participants, but did not affect the accessibility of negative self-concepts. These results suggest that failures would heighten the accessibility of positive self-concepts in humble people but would not affect the accessibility of their negative self-concepts. In contrast, the neutral participants demonstrated both an increase in accessibility of their positive self-concepts, and a decrease in accessibility of their negative self-concepts, in the negative feedback condition relative to the no feedback condition. These findings are consistent with the general strategy of self-enhancement following ego-threats in which positive selfconcepts are emphasized and negative self-concepts rejected (Sedikides & Strube, 1995). The results revealed substantial differences between humility and LSE. Under the negative feedback condition, the humility participants showed a higher level of accessibility of positive self-concept, and a lower level of accessibility of negative selfconcept, than the LSE participants. Also in contrast to the humility participants, and in direct opposite to the neutral participants, the LSE participants showed a higher accessibility of negative self-concepts, and a lower accessibility of positive self-concepts, following negative feedback relative to no feedback. The results suggest that LSE individuals not only tend not to activate positive self-information in a failure situation which humble individuals are more likely to do, they also tend not to activate the selfenhancing strategies typical of the average person when dealing with an ego-threat. The results also indicate that the LSE participants exhibited higher accessibility to negative self-concepts compared to the neutral participants in the no feedback condition, thus 34 suggesting that the arithmetic task alone (without feedback) may be threatening to those feeling low self-esteem. 35 CHAPTER FOUR EXPERIMENT 3 Overview Experiment 3 has two objectives. First, it aimed to test the robustness of the findings in Experiments 2 concerning the difference between the humility and neutral conditions, using the same latency response task. Second, instead of a LSE condition, a high self-esteem (HSE) condition was induced (using the recall method) so as to differentiate the effects of humility and HSE. It is posited that humble people enjoy a good sense of self-worth. However, HSE individuals are also known to have a positive sense of self-worth and it remained a question whether humble and HSE would elicit similar or differences responses to failure. It has been found that HSE tend to respond less negatively to failures as they are able to recruit positive thoughts of themselves following a failure (Dodgson & Wood, 1998). Further, they are more likely to employ compensatory strategies such as discrediting the negative feedback or attributing failures to external environment (Blaine & Crocker, 1993). In contrast, although humble individuals are also aware of their strengths, they are just as likely to be aware of their weaknesses. Hence, they will be less likely to resort to such compensatory strategies to cope with a failure. I hypothesized that after negative feedback, the HSE participants should display the shortest RT to positive traits words, followed by the neutral condition, and the humble participates should demonstrate the longest RT, and the reverse pattern was expected for the negative traits words. In addition, I hypothesized that participants in all three conditions (i.e., HSE, humble and neutral) would again respond faster to the positive trait 36 words when given negative feedback compared to when they were not given any feedback. I predicted that both the HSE and neutral participants would respond slower to the negative trait words following negative feedback relative to no feedback. No prediction was made with regard to the humility participants’ responses to the negative trait words as a function of feedback. Method Participants 133 participants (37 males and 99 females) from NUS took part in the experiment in exchange for partial course credits. The average age of the participants is 20.3 (SD = 1.27). 6 participants were excluded as they failed the manipulation checks (to be described further in Results). The participants were randomly assigned to one of the six conditions in a 3 (condition: neutral versus humble versus HSE) by 2 (feedback: no feedback versus negative feedback) factorial design. Procedure The procedure was the same as that of Experiment 2 except the change in HSE condition. Participants in this condition were asked to recall an incident which lifted their self-esteem. This was defined to them as an incident which made them had highly positive evaluations of themselves. High self-esteem was further defined as feeling successful and having positive aspects of themselves that they feel proud of. As in the humility condition, they were encouraged to describe as much details as possible, and were furnished with similar question prompts. They then went thru the same “Conceptual Thinking Study” and self-concept accessibility task. 37 Measures Reported humility Participants in all conditions rated the same reported humility as in Experiments 1 and 2 (α = .81). Negative feedback Participants in the negative feedback condition were also asked the same percentile question and rated the same perceived feedback accuracy item. State self-esteem The same modified version of state self-esteem scale (Heatherton & Polivy, 1991) was used an in Experiment 2. Results Manipulation Check Induction of Humility Reliability analysis conducted on the manipulation check items revealed a high Cronbach’s α of .81. A one-way ANOVA conducted on the reported humility indicated participants in humble condition rated feeling more humble (M = 4.71, SD =.50) than participants in neutral condition (M = 2.98, SD = .67) and HSE condition (M = 3.57, SD = .75), F(2,127) = 80.74, p < .001, ηp2 = .57. This indicated that the manipulation was successful. Two coders (one of them blind to the research hypothesis) coded the presence/absence of the same six humility attributes in the recall descriptions. A high inter-rater reliability was obtained, intra-class correlation coefficient (ICC) = .85. The attribute proportions in the humility condition were comparable to those obtained in Experiment 1 and 2 (see Table 1). All of the humble attributes except “An assured sense of self-worth” and “Willingness to improve on the imperfection” were significantly above 50%. This implies that at least half of the participants’ recall descriptions contained the other four attributes. In contrast, the humble attributes proportions in the HSE condition were extremely low (< 5%), and also were so low in 38 variance that analyses on them would not be meaningful. I then conducted another onesample t-test comparing the agreed rating against 50% and it was found that the recalled humility descriptions contain at least 50% of the humility traits, t(46) = 5.93, p < .001, ηp2 = .43. Negative Feedback 6 participants recalled their scores to be less than the 20th percentile (1 from the humility condition, 3 from the neutral condition and 2 from the HSE condition). The average rating of accuracy is 5.10 (SD = .98) which ranges from 1 (not accurate at all) to 7 (very accurate). Induction of High self-esteem Participants in HSE condition (M = 3.84, SD = .74) indicated higher ratings of self-esteem than humble (M = 3.02, SD = .61) and neutral condition (M = 2.89, SD = .56), F(2,124) = 26.14, p < .001, ηp2 = .30. Main Analysis A 3 (condition: humble versus neutral versus LSE) × 2 (feedback: negative feedback versus no feedback) × 2 (traits: positive versus negative) mixed ANOVA was conducted on the mean number of self responses (see Experiment 2). No significant main effects of condition, F(2,121) = 1.60, p = .206, ηp2 = .03, and feedback, F(1,121) = 2.76, p = .10, ηp2 = .02 were found. The interactions were also not significant: condition × feedback, F(2,121) = 1.29, p =.28, ηp2 = .02, condition × traits, F(2,121) = .03, p =.97, ηp2 < .001, feedback × traits, F(1,121) = .09, p =.76, ηp2 = .001, and condition × feedback × traits, F(2,121) = 1.23, p =.30, ηp2 = .02.The results indicate that participants across the conditions do not differ in their response tendencies. Again, consistent with past literature it was found that participants RT for self responses were faster (M = 723.6) than nonself responses (M = 819.8). And since the conditions do not differ in their 39 response latency, the difference in RT for self and nonself responses do not pose as a potential artifact. In addition, most participants select self for the positive traits and nonself for the negative traits (an average of 90% of the responses). To examine if the response latency differs significantly across positive and negative traits, I subjected the RT to a 3 (condition: humble versus neutral versus LSE) × 2 (feedback: negative feedback versus no feedback) × 2 (traits: positive versus negative) mixed ANOVA. There is no significant main effect of condition, F(2,121) = 1.59, p =.21, ηp2 = .02. A significant main effect of feedback, F(1,121) = 5.05, p =.03, ηp2 = .04, a significant condition × feedback interaction, F(1,121) = 3.75, p =.03, ηp2 = .06, and a significant condition × feedback × traits interaction, F(1,121) = 3.95, p =.02, ηp2 = .06 were obtained. This implies that the response latency differs significantly between positive and negative traits. As such, the accessibility of positive and negative traits were analyzed and presented separately. The finding that nonself responses are significantly slower than self responses was supported. And because an average of 90% of the responses for the negative traits were nonself and for the positive traits were self, the response latency between positive and negative traits will differs. Accessibility of positive self-concepts A 3 (condition: humble versus neutral versus HSE) × 2 (feedback: negative feedback versus no feedback) ANOVA revealed a significant main effect of condition, F(2,121) = 5.38, p = .01, ηp2 = .08, and a significant main effect of feedback, F(1,121) = 37.74, p < .001, ηp2 = .24. Post-hoc analyses revealed that after given negative feedback, participants in neutral condition responded faster (M = 799.5) to the positive traits than participants in humble condition (M = 908.7), t(40) = 3.65, p = .001, ηp2 = .25 and participants in HSE condition (M = 40 910.0), t(39) = 2.55, p = .02, ηp2 = .14. There is no significant difference between humble and HSE condition t(41) = .22, p = .82, ηp2 < .001 and among the conditions when feedback was not given, F(2,61) = 1.02, p = .37, ηp2 = .03. Further analyses revealed that in all three conditions, participants who received the negative feedback responded faster to the positive traits than those not given any feedback, all ts > 2.32 and ps < .001 (see means in Table 5). Accessibility of negative self-concepts The same analyses conducted revealed a significant main effect of condition, F(2,121) = 4.08, p = .02, ηp2 = .06, and a significant interaction, F(2,121) = 4.78, p = .01, ηp2 = .07. It was also found in negative feedback condition, participants in humble condition RT (M = 927.9) have a shorter RT to the negative traits presented than participants in neutral condition, (M = 1184.9), t(40) = 2.60, p = .01, ηp2 = .14 and participants in HSE condition, (M = 1163.4), t(41) = 2.48, p = .02, ηp2 = .13. However, there is no significant difference between neutral and HSE conditions, t(45) = .39, p = .71, ηp2 < .001.There is also no significant difference among the conditions when participants were not given any feedback, F(2,63) = .13, p = .88, ηp2 = .004 . Further analyses revealed that within the humble condition, there were no significant different between participants who received negative feedback and those without any feedback, t(44) = 1.39, p = .17, ηp2 = .04. Within both neutral and HSE conditions, participants displayed a longer RT when given negative feedback compared to those not given any feedback, t(38)neutral = 2.65, p = .01, ηp2 = .15, and , t(39)HSE = 2.81, p = .008, ηp2 = .17 (see means in Table 5). 41 Discussion Experiment 3 largely replicated the findings of Experiment 2 with regard to the differences between the humility and neutral conditions. Following negative feedback, the humility participants showed lower accessibility of positive self-concepts, but higher accessibility of negative self-concepts, than the neutral participants. As in Experiment 2, the failure feedback increased the accessibility of positive self-concepts and decreased the accessibility of negative self-concepts among the neutral participants relative to no feedback. Also replicating Experiment 2, the failure feedback increased the accessibility of positive self-concepts among the humility participants relative to no feedback, but did not affect the accessibility of their negative self-concepts. Overall, the results support the current contention that humility mutes the self-enhancing, compensatory mechanism typically activated under neutral conditions in response to ego-threats. The results also indicated significant differences between humility and HSE participants. After negative feedback, humility participants exhibited higher accessibility to negative self-concepts than HSE participants as predicted, though, there were no significant differences between them in terms of the accessibility of positive selfconcepts. The accessibility of positive self-concepts increased in the feedback condition relative to no feedback for both the humility and HSE participants, but it was only the HSE participants who demonstrated weaker accessibility of negative self-concepts. These results indicate that humility should also not be equated with HSE. 42 CHAPTER FIVE GENERAL DISCUSSION In the current research, I posited that humility enables non-defensive response to failures. Existing findings suggest that humility is associated with an assured sense of self-worth and a willingness to acknowledge and accept one’s shortcomings. Drawing from this literature, I proposed that individuals in a state of humility are less likely to cope with failures using self-enhancing strategies. More specifically, whereas people tend to respond to by failures by emphasizing their strengths and rejecting their weaknesses, humble individuals are less likely to magnify their strengths or diminish their weaknesses. Therefore, I hypothesized that humility should lead to a weaker activation of positive self-concepts and stronger activation of negative self-concepts compared to the neutral state. Promising evidence was obtained that supported the hypothesis. Across three experiments, humility was induced by having participants recalled personal incidents of humility. There is good evidence from the manipulation checks and coding of the recall descriptions that this method was effective in inducing participants to feel humble. In Experiment 1, a self-report listing method was used to measure activation of positive and negative self-concepts. It was found that participants induced with humility listed a significantly lower proportion of positive self-traits than those in the neutral condition after receiving a negative feedback. This finding was conceptually replicated in Experiments 2 and 3. Using a response latency paradigm in both experiments, it was found that after receiving negative feedback, neutral condition responded faster to the positive self-concepts than humble condition. This pattern was not 43 found when participants were not given any negative feedback. A reversed pattern was found for the negative self-concepts such that while the neutral condition displayed a longer RT than humble participants. Within both the humble and neutral conditions, participants displayed heightened accessibility (i.e., shorter RT) to the positive selfconcepts after receiving negative feedback than when they were not given any feedback. A reverse pattern was found for the negative self-concepts in the neutral condition such that after receiving negative feedback, they reacted slower to the negative self-concepts than when they did not receive any feedback. This was not the case for humble condition which did not show any differences in the RT of the negative self-concepts regardless of negative feedback. To demonstrate the conceptual differences between humility with LSE and HSE, the conditions were added to Experiment2 and 3 respectively. In Experiment 2 after receiving negative feedback, the LSE condition showed the longest RT to positive selfconcepts among the three conditions. In addition, they also displayed a significantly shorter RT to the negative self-concepts than both humble and neutral conditions even without any negative feedback. Within the condition, after receiving negative feedback, they responded slower to the positive self-concepts compared to no feedback given. In Experiment 3 with a change of high self-esteem (HSE) condition, it was found that after receiving negative feedback, neutral condition responded faster to the positive selfconcepts than both humble and HSE conditions. However, both HSE and neutral conditions showed longer RT to the negative self-concepts compared to humble condition. These were not found when participants were not given any negative feedback. Within all three conditions, participants who received the negative feedback responded faster to the 44 positive traits than those not given any feedback. For the negative self-concepts, within both neutral and HSE conditions those received negative feedback have a longer RT than those without feedback. This was not found for the humble condition. Reconciling the findings with Literature Self-enhancing is a common strategy to cope with failures (Feick & Rhodewalt, 1997; Sedikides & Strube, 1995; Arkin, 1981). This strategy includes people undermining their failures by reminding themselves of their positive traits (Baumeister & Jones, 1978; Steele, 1988). This coping strategy was clearly displayed in the neutral condition. After negative feedback, the neutral condition participants displayed heightened accessibility of positive self-concepts and decreased accessibility of negative self-concepts. Humble individuals, on the other hand, displayed such tendencies to a significantly lesser degree. Compared to the neutral conditions participants, they demonstrated weaker activation of positive self-concepts and stronger activation of negative self-concepts. These findings appear consistent with existing profile of humble individuals which construes them as less prone to brag about their strengths and downplay their weaknesses in the presence of an ego-threat (Sandage, 1999; Emmons, 1999; Jennings et al., 2005). The findings are also consistent with a view expounded by Carl Rogers (1961) that secured individuals are able to experience and accept every aspects of themselves, both the good and bad, and feel less compelled to continually seek the approval of others. Humble individuals are known to display qualities associated with secure individuals, including an affirmed sense of self-worth and less defensive approach in reacting to personal affronts. The current findings provide further empirical indications 45 that humble individuals are high self-acceptance and feel less need to protect their identity or self-esteem when attacked. After the negative feedback, LSE individuals responded slower to the positive self-concepts (i.e., longer RT) than the neutral participants. Yet, a reverse pattern was seen for the negative self-concepts; LSE responded quicker than neutral participants. The findings in the LSE condition are consistent with the literature which supports the notion that individuals with low self-esteem tend not to engage in self-enhancing strategies following failures (Baumeister, 1982). Further, they are also known to overgeneralize their weaknesses in response to failures (Kernis et al., 1989; Brown & Dutton, 1995). It is notable that even without negative feedback, the LSE condition participants displayed a heightened accessibility of negative self-concepts. This implies that the arithmetic task itself poses as a threat, and may also suggests that LSE is associated a chronic heightened activation of self-concepts under neutral conditions with no ego-threat whatsoever. HSE individuals are considered to have a secured sense of self-esteem and past research have shown that individuals with healthy self-esteem are more able to focus on their strengths as an adaptive response to life’s challenges like a failure (Brown & Smart, 1991; Spencer et al., 1993; Dodgson & Wood, 1998). This was demonstrated in the current research. When HSE participants were given negative feedback, they displayed a heightened accessibility of positive self-concepts (i.e., shorter RT). Studies have also found that individuals with HSE are particularly prone to engage in compensatory strategies such after a failure (Wood, Giordano-Beech, Taylor, Michela, & Gaus, 1994), probably more so than humble individuals. Consistent findings were found in Experiment 46 3; after receiving negative feedback HSE participants displayed a diminished accessibility of negative self-concepts compared to humble participants. Though the current method of failure manipulation has been demonstrated to successfully induce feelings of failure (e.g., Dodgson & Wood, 1998), I do acknowledge that having a negative feedback do not equate to experience of failure. However the current findings seem to suggest that participants do experience failure (hence the increase accessibility to positive self-concepts and the converse diminished accessibility to negative self-concepts). Nonetheless, further studies can explore using an alternate method of failure manipulation. Further Implications and Conclusion The idea that the trait humility bestows positive life outcomes has been consistently suggested (Emmons, 1999; Exline & Geyer, 2004; Petersen & Seligman, 2004; Tangney, 2002). For instance, humble employees often demonstrate willingness to learn and responding positively to negative feedback (Reave, 2005) which will benefit their interpersonal relationships. Correlations have also been found between humility and good academic results (Johnson, Rowatt, & Petrini, 2011; Rowatt et al., 2006). This character strength helps one to adapt to life adversaries. As demonstrated in the current research, humble individuals are able to take failures in their stride. Having an assured sense of self help the humble to focus on their strengths in order to cope with the ego threat. Yet, this focus is not an inflation of positive self-views. Further, they are willing to accept their shortcomings and are not compel to reject negative self-information to disconfirm the undesirable failure feedback. 47 Failures are inevitable in life and they affect some more than others. Humble individuals are particularly not as affected by personal failures. Being able to accept one’s limitations will oriented one towards learning and growth. This might suggests that humble individuals have a higher subjective wellbeing than the non-humble people. These predictive associations can be studied in future research. As noted earlier, humility is a dynamic and possibly multi-dimensional construct. Furthermore, the potential benefits of humility make this an important construct to examine. However, the lack of a suitable measurement tool combined with the richness of this construct makes it difficult to study. And because of this lack of theory-based, reliable and valid measure, empirical research on humility remained limited. There is a need to sharpen the measurement technique for which the field can be advanced. With a reliable and valid measurement tool, the current results can be replicated. In the current research, the recall method proves to be an effective way of humility induction but this method does have its cons. Though definition of humility was given in the process of doing a situation recall, participants can still hold a mistaken view of the trait. Further, dispositional humility cannot be assessed with the current method as well. Nonetheless, though the risk of having a mistaken view of humility exists, steps have been taken to minimize this possibility. For instance, all recall descriptions were coded to check the recall content coincides with what is known regarding humility. Humility is more than a personality trait; it is a character strength that influences how we cope with failures. Indeed, the three experiments support the notion that being humble enables one to draw on his or her strengths to buffer against failures. Yet, these strengths are not falsely inflated. Having an assured sense of self-worth means that these 48 humble individuals do not shun negative feedback, rather they would acknowledge and accept their weaknesses. 49 Table 1 Percentage of Humility Traits in Participants Recall Descriptions (Experiments 1 to 3) Experiment 1 (ICC = .84) Rater 1 Rater Agreed 2 rating Experiment 2 (ICC = .93) Rater Rater Agreed 1 2 rating Experiment 3 (ICC = .85) Rater Rater Agreed 1 2 rating List of Humility Traits 1. An assured sense of selfworth (i.e., an accurate sense of one’s strengths and weaknesses) 53.1 56.3 56.3* 52.4 50.8 52.* 41.7 45.0 45.0 2. Tendency to keep one’s abilities and accomplishments in perspective 56.3 59.4 59.4* 63.5 58.7 58.7* 63.3 60.0 63.3* 3. Willingness to acknowledge and accept one’s mistakes and imperfections 65.6 59.4 65.6* 63.5 61.9 63.5* 53.3 56.7 56.7* 4. Lack of self-deprecating tendencies 46.9 50.0 50.0 53.9 49.2 53.9* 60.0 66.7 66.7* 5. Free from arrogance 62.5 59.4 59.4* 68.3 71.4 71.4* 70.0 65.0 70.0* 6. Willingness to improve on the imperfection 59.4 53.1 53.1* 50.8 63.5 63.5* 51.7 50.0 50.0 Average % of traits in recall 52.7 description * percentage significantly above 50%, p < .001 54.6 53.7 62.0 56.2 60.1 52.3 52.2 58.1 50 Table 2 Proportion of Positive Traits listed across conditions (Experiment 1) Humble Negative No Feedback Feedback (n = 17) (n = 15) Proportion of Positive Traits Listed M .60 .49 SD .14 .12 Neutral Negative No Feedback Feedback (n = 18) (n = 13) .68 .19 .58 .20 Number of Positive Traits Listed M SD 8.18 2.56 5.67 2.13 10.0 5.11 9.30 4.17 Number of Negative Traits Listed M SD 5.82 3.04 5.53 1.55 4.17 2.00 5.46 1.57 51 Table 3 Response Tendencies: Mean Number of self versus nonself selections across conditions (Experiment 2 & 3) Experiment 2 Humble Neutral LSE Positive-self NegativePositive-self NegativePositive-self Negativenonself nonself nonself Condition M SD M SD M SD M SD M SD M SD Negative 35.39 3.72 34.0 5.69 36.0 3.68 33.10 5.88 36.0 3.33 33.47 7.16 Feedback No feedback 36.25 Condition Negative Feedback 3.43 33.95 7.06 Humble Positive-self Negativenonself M SD M SD 36.31 3.67 33.22 7.25 No feedback 35.0 4.80 32.29 Note that means are out of 40 trials 5.14 36.78 3.32 33.56 5.76 36.13 3.47 32.0 6.92 Experiment 3 Neutral Positive-self Negativenonself M SD M SD 34.55 6.24 31.75 7.14 M 35.71 SD 3.91 Negativenonself M SD 32.62 5.12 34.30 35.40 4.42 32.85 5.26 35.55 4.26 HSE Positive-self 3.91 52 Table 4 Means of Reaction Time to Positive and Negative Self-concepts across conditions (Experiment 2) Humble Neutral Negative No Negative No Feedback Feedback Feedback Feedback (n = 31) (n = 24) (n = 31) (n = 27) Positive Self-concepts Reaction Time (milliseconds) M 931.1 1026.2 836.1 1014.4 SD 166.4 125.8 132.7 160.4 LSE Negative Feedback (n = 32) No Feedback (n = 31) 1201.8 223.6 1067.6 147.6 867.4 177.2 963.5 154.3 Negative Self-concepts Reaction Time (milliseconds) M SD 1005.7 166.8 1078.2 168.6 1210.1 136.2 1090.5 208.7 53 Table 5 Means of Reaction Time to Positive and Negative Self-concepts across conditions (Experiment 3) Humble Neutral Negative No Negative No Feedback Feedback Feedback Feedback (n = 22) (n = 24) (n = 20) (n = 20) Positive Self-concepts Reaction Time (milliseconds) M 908.7 1070.1 799.5 994.2 SD 71.6 166.5 112.1 151.36 HSE Negative Feedback (n = 21) No Feedback (n = 20) 910.0 154.3 1043.6 198.3 1163.4 168.9 1000.7 198.3 Negative Self-concepts Reaction Time (milliseconds) M SD 927.9 231.8 1032.5 218.7 1184.9 175.7 1028.4 183.9 54 References Adler, A. (1956). The Individual Psychology of Alfred Adler. H. L. Ansbacher and R. R. Ansbacher (Eds.). New York: Harper Torchbooks. Arkin, R. M. (1981). Self-presentation styles. In J. Tedeschi (Ed.), Impression management theory and research (pp. 311-333). New York: Academic Press. Asendorpf, J. B., & Ostendorf, F. (1998). Is self-enhancement healthy? Conceptual psychometric, and empirical analysis. Journal of Personality and Social Psychology, 74, 955-966. doi:10.1037//0022-3514.74.4.955 Ashton, M., & Lee, K. (2005). Honesty-Humility, the Big Five, and the Five-Factor Model. Journal of Personality, 73, 1321-1353. doi:10.1016/j.paid.2004.09.016 Ashton, M., & Lee, K. (2007). Empirical, theoretical, and practical advantages of the HEXACO model of personality structure. Personality and Social Psychology Review, 11, 150-166. doi:10.1177/1088868306294907 Ashton, M., & Lee, K. (2008). The prediction of Honesty-Humility-related criteria by the HEXACO and Five-Factor Models of personality. Journal of Research in Personality, 42, 1216-1228. doi:10.1016/j.jrp.2008.03.006 Baumeister, R. E, & Jones, E. E. (1978). When self-perception is constrained by the target's knowledge: Consistency and compensation. Journal of Personality and Social Psychology, 36, 608-618. doi:10.1037//0022-3514.36.6.608. Baumeister, R. F. (1982). Self-esteem, self-presentation, and future interaction: A dilemma of reputation. Journal of Personality, 50, 29-45. doi:10.1111/j.14676494.1982.tb00743.x 55 Blaine, B., & Crocker, J. (1993). Self-esteem and self-serving biases in reactions to positive and negative events: An integrative review. In R.F. Baumeister (Ed.), Self-esteem: The puzzle of low self-regard (pp. 55-85). New York: Plenum. Brockner, J. (1979). The effects of self-esteem, success–failure, and self-consciousness on task performance. Journal of Personality, 37, 1732-1741. doi: 10.1037/00223514.37.10.1732 Brown, J.D., & Smart, S.A. (1991). The self and social conduct: Linking selfrepresentations to prosocial behavior. Journal of Personality and Social Psychology, 60, 368-375. doi:10.1037/0022-3514.60.3.368 Brown, J. D., & Dutton, K. A. (1995). The thrill of victory, the complexity of defeat: Self-esteem and people's emotional reactions to success and failure. Journal of Personality and Social Psychology, 68, 712- 722. doi:10.1037//00223514.68.4.712 Clark, A. T. (1992). Humility. In D. H. Ludlow (Ed.), Encyclopedia of Mormonism (pp. 663-664). New York: Macmillan. Compton, W. C., (2001). Toward a tripartite factor structure of mental health: Subjective well-being, personal growth, and religiosity. The Journal of Psychology, 135, 486-500. doi:10.1080/00223980109603714 Collins, J. (2001). Good to Great. New York:Harper, Collins. Colvin, C. R., Block, J. B., & Funder, D. C. (1995). Overly positive self-evaluations and personality: Negative implications for mental health. Journal of Personality and Social Psychology, 68, 1152-1162. doi:10.1037//0022-3514.68.6.1152 56 Davis, D. E., Worthington, E. L., Jr., & Hook, J. N. (2010). Relational humility: A review of definitions and measurement strategies. Journal of Positive Psychology, 5, 243252. http://dx.doi.org/10.1080/17439761003791672 Dodgson, P.G., & Wood, J.V. (1998). Self-esteem and the cognitive accessibility of strengths and weaknesses after failure. Journal of Personality and Social Psychology, 75, 178–197. doi:10.1037//0022-3514.75.1.178 Emmons, R. A. (1999). The psychology of ultimate concerns. New York: Guilford. Exline, J. J., Bushman, B., Faber, J., & Phillips, C. (2000). Pride gets in the way: Selfprotection works against forgiveness. In J. J. Exline (Chair), Ouch! Who said forgiveness was easy? Symposium conducted at the annual meeting of the Society for Personality and Social Psychology, Nashville, TN. Exline, J. J., Baumeister, R. F., Bushman, B. J., Campbell, W. K., & Finkel, E. J. (2004). Too proud to let go: Narcissistic entitlement as a barrier to forgiveness. Journal of Personality and Social Psychology, 87, 894-912. . doi:10.1037/00223514.87.6.894 Exline, J. J., & Geyer, A. L. (2004). Perceptions of humility: A preliminary study. Self and Identity, 3, 95-114. Doi: 10.1080/13576500342000077 Fazio, R. H. (1990). A practical guide to the use of response latency in social psychological research. In C. Hendrick & M. S. Clark (Eds.), Review of personality and social psychology: Vol. 11. Research methods in personality and social psychology (pp. 74- 97 ). Newbury Park, CA: Sage. 57 Feick, D. L., & Rhodewalt, F. (1997). The double-edged sword of self-handicapping: Discounting, augmentation, and the protection and enhancement of self-esteem. Motivation and Emotion, 21, 147-163. doi:10.1023/A:1024434600296 Fontana, A.F., Rosenberg, R.L., Burg, M.M. Kerns, R.D. & Colonese, K.L. (1990). Type A behavior and self referencing: Interactive risk factors? Journal of Social Behavior and Personality, 5, 215-232. Gerrards-Hesse, A., Spies, K., & Hesse, F. W. (1994). Experimental inductions of emotional states and their effectiveness: A review. British Journal of Psychology, 85, 55-78. doi:10.1111/j.2044-8295.1994.tb02508.x Gibson, B. (2007). The role of individual differences in sandbagging on selective avoidance of self-evaluative information. Journal of Research in Personality, 41, 481-487. doi:10.1016/j.jrp.2006.04.006, Greenberg, J., & Pyszczynski, T. (1986). Persistent high self-focus after failure and low self-focus after success: the depressive self-focusing style. J Pers Soc Psychol, 50(5), 1039-1044. doi:10.1037//0022-3514.50.5.1039 Halling, S., Kunz, G., & Rowe, J. O. (1994). The contributions of dialogal psychology to phenomenological research. Journal of Humanistic Psychology, 34, 109-131. doi:10.1177/00221678940341007 Hareli, S., & Weiner, B. (2000). Accounts for success as determinants of perceived arrogance and modesty. Motivation and Emotion, 17, 215-236. doi:10.1023/A:1005666212320 58 Harmon-Jones, E., Peterson, H., & Vaughn, K. (2003). The dissonance-inducing effects of an inconsistency between experienced empathy and knowledge of past failures to help: Support for the action-based model of dissonance. Basic and Applied Social Psychology, 25, 69-78. doi:10.1207/S15324834BASP2501_5 Heatherton, T. F., & Polivy, J. (1991). Development and validation of a scale for measuring state self-esteem. Journal of Personality and Social Psychology, 60, 895–910. doi:10.1037//0022-3514.60.6.895 Higgins, E. T. (1996). Knowledge activation: Accessibility, applicability, and salience. In E. T. Higgins and A. W. Kruglanski (Eds.), Social psychology: Handbook of basic principles (pp. 133 – 168). New York: Guilford Press. Hineline, P. N. (1992). A Self-interpretive behavior analysis. American Psychologist , 47 , 1274-1286. doi:10.1037//0003-066X.47.11.1274 Hwang, C. (1982). Studies in Chinese personality: A critical review. Bulletin of Educational Psychology, 15, 227-242. Jennings, L., Sovereign, A., Bottorff, N., Mussell, M. P., & Vye, C. (2005). Nine ethical values of master therapists. Journal of Mental Health Counseling, 27, 32-47. Johnson, M. K., Rowatt, W. C., & Petrini, L. (2011). A new trait on the market: Honestyhumility as a unique predictor of job performance. Personality and Individual Differences, 50, 857-862. http://dx.doi.org/10.1016/j.paid.2011.01.011 Kernis, M, H., Brockner, J., & Frankel, B. S. (1989). Self-esteem and reactions to failure: The mediating role of overgeneralization. Journal of Personality and Social Psychology, 57, 707-714. doi:10.1037//0022-3514.57.4.707 Kurtz, E., & Ketcham, K. (1992). The spirituality of imperfection. New York: Bantam. 59 Lawler, K. A., Younger, J. W., Piferi, R. L., Billington, E., Jobe, R., Edmondson, K., & Jones, W. H. (2003). A change of heart: Cardiovascular correlates of forgiveness in response to interpersonal conflict. Journal of Behavioral Medicine, 26, 373393. doi:10.1023/A:1025771716686 Lawler, K. A., Younger, J. W., Piferi, R. L., Jobe., Edmondson, K. A., & Jones, W. H. (2005). The unique effects of forgiveness on health: An exploration of pathways. Journal of Behavioral Medicine, 28, 157-167. doi:10.1007/s10865-005-3665-2 McFarlin, D. B., & Blascovich, J. (1981). Effects of self-esteem and performance feedback on future affective preferences and cognitive expectations. Journal of Personality and Social Psychology, 40, 521 - 531. doi:10.1037//00223514.40.3.521 Markus, H., & Kunda, Z. (1986). Stability and malleability of the self-concept. Journal of Personality and Social Psychology, 51, 858-866. . doi:10.1037//00223514.51.4.858 Means, J. R., Wilson, G. L., Sturm, C., & Biron, J. E. (1990). Humility as a psychotherapeutic formulation. Counseling Psychology Quarterly 3, 211-215. doi:10.1080/09515079008254249 Moore, D.A. (2007). Not so above average after all: When people believe they are worse than average and its implications for theories of bias in social comparison. Organizational Behavior and Human Decision Processes, 102, 42-58. doi:10.1016/j.obhdp.2006.09.005 Morrison, S. M. (1979). The effects of success and failure on self-esteem. Australian Journal of Psychology, 31, 1-8. doi: 10.1080/00049537908254643 60 Myers, D. G. (1979). The inflated self: Human illusions and the biblical call to hope. New York: Seabury. Myers, D. G. (1995). Humility: Theology meets psychology. Reformed Review, 48, 195206. Myers, D. G. (2000). The psychology of humility. In R. L. Herrmann (Ed.), God, science, and humility (pp. 135-175). London: Templeton Foundation. Neff, K. D., Hsieh, Y., & Dejitterat, K. (2005). Self-compassion, achievement goals, and coping with academic failure. Self and Identity, 4, 263−287. doi:10.1080/13576500444000317 Nurius, P. S., & Markus, H. (1990). Situational variability in the self-concept: Appraisals, expectancies, and asymmetries. Journal of Social and Clinical Psychology, 9, 316-333. doi:10.1521/jscp.1990.9.3.316 Peterson, C, & Seligman, M. E. P. (2004). Character strengths and virtues: A handbook and classification. Washington, DC: American Psychological Association. Peters, A.S., Rowatt, W.C., & Johnson, M.K. (2011). Associations between dispositional humility and social relationship quality. Psychology, 2, 155-161. doi:10.4236/psych.2011.23025 Powers, C., Nam, R. K., Rowatt, W. C., & Hill, P. C. (2007) Associations between humility, spiritual transcendence and forgiveness. Research in the Social Scientific Study of Religion, 18, 75-94. http://dx.doi.org/10.1163/ej.9789004158511.i-301.32 Reave, L. (2005). Spiritual values and practices related to leadership effectiveness. The Leadership Quarterly, 16(5), 655-687. doi:10.1016/j.leaqua.2005.07.003 61 Richards, N. (1992). Humility. Philadelphia: Temple University Press Roberts, R. C. (1983). Spirituality and Human Emotion. Grand Rapids, MI: Wm. B. Eerdmans Publishing Robins, R.W., & Beer, J.S. (2001). Positive illusions about the self: Short-term benefits and long-term costs. Journal of Personality and Social Psychology, 80, 340-352. doi:10.1037//0022-3514.80.2.340 Robinson, M. D., Johnson, J. T., & Shields, S. A. (1995). On the advantages of modesty: The benefits of a balanced self-presentation. Communication Research, 22(5), 575-591. doi:10.1177/009365095022005003 Rogers, C. R. (1961). On becoming a person. Boston: Houghton Mifflin. Rowatt. W. C, Powers, C, Targhetta, V., Comer, J., Kennedy, S., & Labouff, J. (2006). Development and initial validation of an implicit measure of humility relative to arrogance. The Journal of Positive Psychology, 7(4), 198-211. doi:10.1080/17439760600885671 Sandage, S. J. (1999). An ego-humility model of forgiveness: Theoretical foundations. Marriage and Family: A Christian Journal, 2, 259 – 276. Sandage, S. J., & Wiens, T. W. (2001). Contextualizing models of humility and forgiveness: A reply to Gassin. Journal of Psychology and Theology, 29, 201-211. Sedikides, C, & Strube, M. J. (1997). Self-evaluation: To thine own self be good, to thine own self be sure, to thine own self be true, and to thine own self be better. In M. P. Zanna (Ed.), Advances in experimental social psychology, Vol. 29 (pp. 209-269). San Diego, CA: Academic Press. 62 Sedikides, C., Gregg., A. P., & Hart, C. M. (2007). The importance of being modest. In C. Sedikides & S. Spencer (Eds.), The self: Frontiers in social psychology (pp. 163184). New York: Psychology Press. Shrauger, J. S., & Rosenberg, S. E. (1970). Self-esteem and the effects of success and failure feedback on performance. Journal of Personality, 38, 404-417. doi:10.1111/j.1467-6494.1970.tb00018.x Shrauger, J. S. (1975). Responses to evaluation as a function of initial self-perceptions. Psychological Bulletin, 82, 581-596. doi:10.1037/h0076791 Spencer, S. J., Josephs, R. A., & Steele, C. M. (1993). Low self-esteem: The uphill struggle for self-integrity. In R. F. Baumeister (Ed.), Self-esteem: The puzzle of low self-regard (pp. 21-36). New York:Plenum. Steele, C.M. (1988). The psychology of self-affirmation: Sustaining the integrity of the self. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 21, pp. 261-302). San Diego, CA: Academic Press. Steele, C. M., Spencer, S. J., & Lynch, M. (1993). Self-image and dissonance: The role of affirmational resources. Journal of Personality and Social Psychology, 64, 885896. doi:10.1037//0022-3514.64.6.885 Tabachnick, B.G., & Fidell, L.S. (2001). Using multivariate statistics (4th ed.). Boston: Allyn & Bacon. Tangney, J. P. (2000). Humility: Theoretical perspectives, empirical findings and directions for future research. Journal of Social and Clinical Psychology, 19(1), 70-82. doi:10.1521/jscp.2000.19.1.70 63 Tangney, J. P. (2002). Humility. In C. R. Snyder & S. J. Lopez (Eds.), Handbook of positive psychology (pp. 411-419). London: Oxford University Press. Tangney, J. P. (2005). Humility. In C. R. Snyder & S. J. Lopez (Eds.), Handbook of positive psychology (pp. 411-419). New York: Oxford University Press. Templeton, J. M. (1998). The humble approach. Radnor, PA: Templeton Press. Winkielman, P., Berridge, K. C., & Wilbarger, J. L. (2005). Unconscious affective reactions to masked happy versus angry faces influence consumption behavior and judgments of value. Personality and Social Psychology Bulletin, 31, 121-135. doi:10.1177/0146167204271309 Wood, J.V., Giordano-Beech, M., & Ducharme, M.J. (1999) . Compensating for failure through social comparison. Personality and Social Psychology Bulletin, 25(11), 1370-1386. doi:10.1177/0146167299259004 Wood, J. V., Giordano-Beech, M., Taylor, K. L., Michela, J., & Gaus, V. (1994). Strategies of social comparison among people with low self-esteem: Selfprotection and self-enhancement. Journal of Personality and Social Psychology, 67, 713-731. doi:10.1037/0022-3514.67.4.713 Worthington, E. L. (1998). An empathy-humility-commitment model of forgiveness applied within family dyads. Journal of Family Therapy, 20, 59-76. Yancey, P. (2000). Humility's many faces. Christianity Today, 44, 96. Yang, Z., & Tong, E. M. W. (2010). The effects of subliminal anger and sadness primes on agency appraisals. Emotion, 10, 915-922. doi: 10.1037/a0020306 [...]... commuting to the school and their daily routine on the evening before they go to bed The purpose of getting the neutral participants to describe two activities was to make comparable the time both the humility and neutral participants spent on this recall task After the recall, participants proceeded to a task that measured the current dependent variable They were told that they would be participating in a. .. of the self-concept activated in response to a failure situation between humble and neutral (non -humble) people To simulate a failure situation, some participants were asked to do an arithmetic task and were given feedback that they had done poorly Other participants completed the same task but were not given any feedback Self-concepts were then measured Note that this research examined only global... for the items were averaged to produce reported humility (α = 90) Negative feedback As manipulation checks for the negative feedback manipulation, participants in the negative feedback condition were asked to indicate the percentile they scored They also rated a perceived feedback accuracy item, “How 16 accurate do you perceive the score to be?” on a seven-point scale ranging from 1 (not accurate at all)... with a trait and they had to indicate whether the trait was representative of them by pressing specific keys (one marked as self and the 25 other marked as non- self”) on the button box Detailed instructions were given to emphasize the importance of responding as quickly as possible while being accurate with their response Participants were asked to place their left and right index finger on the specific... et al., 2004) Note also that humble individuals are aware of their strengths These strengths can help them to buffer against failures in life It is thus possible that during failures, humble individuals would draw on their strengths and would be less averse to the negative feedback about themselves, less likely to reject the negative self-information, and more likely to learn from the experience (Tangney,... of their undergraduate cohort Those in the no feedback condition were not told that their scores could be available and were not informed of their results Next, all participants completed the trait-listing task They were told to list down as many positive and negative traits about themselves they could bring to mind at the moment The trait-listing task was self-paced with no time limit Last, the participants... be relatively stable in face of laboratory or academic failures (Heatherton & Polivy, 1991) Also, this factor is irrelevant to the current failure design of the experiment Self-concept Accessibility Task Using frequency of appearance, the traits given by participants from Experiment 1 were ranked Based on the ranking, I generated a list of 40 traits (20 positive and 20 negative) The positive traits... should contained at least 50% of the humble attributes, I conducted another onesample t-test comparing the agreed rating against 50% (i.e., at least 3 out of 6 of the humble attributes) and found that the recall descriptions contained a significantly higher number of humble attributes, t(31) = 2.88, p = 007, ηp2 = 21 Note that there was no data to compare these proportions with because the neutral condition... which affected their selfesteem in a negative way This was defined to them as an incident which made them had negative evaluations of themselves and low self-esteem was further defined as a state of worthlessness As in the humility and neutral conditions, they were encouraged to describe as much details as possible, and were furnished with similar question prompts They then completed the “Conceptual Thinking... and disagreement about the meaning of each attribute was first discussed and resolved The coders then proceeded with a few descriptions as practice It was highly unlikely that any participants would so 18 clearly state that, for example, he/she had a sense of self-acceptance The coders would need to read the entire description and make an inference of whether, in this example, the participant had exhibited ... response to a failure situation between humble and neutral (non -humble) people To simulate a failure situation, some participants were asked to an arithmetic task and were given feedback that they had... getting the neutral participants to describe two activities was to make comparable the time both the humility and neutral participants spent on this recall task After the recall, participants proceeded... accuracy ranged from (not accurate at all) to (very accurate), this suggests that the negative feedback participants in general found the feedback to be a moderately accurate A t-test revealed no